Logo do repositório
Comunidades e Coleções
Todo o repositório
Estatísticas
Sobre
English
العربية
বাংলা
Català
Čeština
Deutsch
Ελληνικά
Español
Suomi
Français
Gàidhlig
ગુજરાતી
हिंदी
Magyar
Italiano
Қазақ
Latviešu
मराठी
Nederlands
Polski
Português
Português do Brasil
Русский
Srpski (lat)
Српски
Svenska
Türkçe
Yкраї́нська
Tiếng Việt
Esqueceu sua senha?
  1. Início
  2. Pesquisar por Data

Navegando por Data de Publicação, começando com "2013-05-24"

Filtrar resultados por ano ou mês
Agora exibindo 1 - 3 de 3
  • Resultados por página
  • Opções de Ordenação
  • Carregando...
    Imagem de Miniatura
    listelement.badge.dso-typeItem,
    Reconhecimento de gestos da Língua Brasileira de Sinais através de máquinas de vetores de suporte e campos aleatórios condicionais ocultos
    (Universidade Federal de São Carlos, 2013-05-24) Souza, César Roberto de; Pizzolato, Ednaldo Brigante; https://lattes.cnpq.br/2821982735490884; https://lattes.cnpq.br/2913925874705853
    This work investigates the use of Support Vector Machines and Hidden Conditional Random Fields in the recognition of signs from the Brazilian Sign Language (Língua Brasileira de Sinais, Libras). Employing basic concepts from Vapnik’s Statistical Learning Theory, the gesture recognition problem is cast as a supervised learning problem defined over images and image streams, avoiding the inherent ill-conditioning present in many density estimation problems through the use of discriminative classification models. From linguistic studies on the structural formation of the Libras sign, a two-layer recognition architecture has been created to operate over features extracted from depth images captured through a depth sensor. This work utilizes quantitative approaches for performance assessment, performing comparisons through contingency tables and statistical hypothesis tests; revealing statistically significant results favoring the aforementioned choice of classification models. Results have shown how the multiclass SVMs organized in Directed Acyclic Graphs provided a needed balance between efficiency and accuracy in the classification of the sub-lexical structures of the Libras, whereas Hidden Conditional Random Fields boosted the system’s recognition rates without for this sacrificing its generalization for unobserved instances.
  • Carregando...
    Imagem de Miniatura
    listelement.badge.dso-typeItem,
    Propriedades de armazenagem de hidrogênio de materiais à base de magnésio e ferro preparados por moagem de alta energia e técnicas de deformação plástica severa
    (Universidade Federal de São Carlos, 2013-05-24) Asselli, Alexandre Augusto Cesario; Botta Filho, Walter José; https://lattes.cnpq.br/8956458007749112; https://lattes.cnpq.br/6693010062191246
    In the doctoral research, the hydrogen storage properties of magnesiumiron based materials prepared by high energy ball milling and techniques of severe plastic deformation were studied. The materials were processed by ball milling under inert atmosphere and reactive milling under hydrogen pressure. The techniques of high pressure torsion, cold forging and rolling were used to prepare the materials in air. Several compositions of Mg and Fe reactants were used to study the behavior of hydrogen absorption and desorption of Mg2FeH6 and MgH2 based materials. The results showed that high energy ball milling was effective to synthesize Mg2FeH6, nevertheless, long milling times (> 12 h) were required and the sample were easily contaminated when exposed to air. Between the techniques of severe plastic deformation, the best results in terms of hydrogen absorption kinetics and capacity were achieved by cold rolling. These results were similar to the milled samples ones, however, cold rolling was carried out in air and with a processing time shorter than 1 minute. The formation mechanism of Mg2FeH6 was studied by measurements of hydrogen absorption kinetics and microstructural characterization through the techniques of X-ray diffraction and scanning and transmission electron microscopy. These analyses allowed to determine that Mg2FeH6 was formed with a columnar morphology by a diffusional process during the hydrogen thermal absorption. The effects of the addition of expanded natural graphite were evaluated to the samples prepared by high energy ball milling and cold rolling. The data showed that the use of this additive resulted in faster kinetics of hydrogen absorption and desorption of the magnesium-iron based materials.
  • Carregando...
    Imagem de Miniatura
    listelement.badge.dso-typeItem,
    Análise proteômica diferencial de proteínas superficiais da membrana de Xanthomonas spp. em interação com hospedeiro cítrico
    (Universidade Federal de São Carlos, 2013-05-24) Carnielli, Carolina Moretto; Novo-Mansur, Maria Teresa Marques; https://lattes.cnpq.br/1198926223396748; https://lattes.cnpq.br/8102669086332844
    The citrus canker is an economically important disease for citrus crop. At the moment, there is no effective means of prevention or cure for this disease, which has contributed to citrus canker wide distribution around the world. The etiologic agents are bacteria of the genus Xanthomonas classified into two species, X. citri and X. fuscans. This study aimed to perform the differential proteomic analysis of cell surface proteins of Xanthomonas citri subsp. citri (XAC), the more virulent specie, between infectious (in vivo) and non-infectious (in vitro) conditions of growth. Additionally, the same analysis was performed by shotgun for XAC against the Xanthomonas fuscans subsp. aurantifolii type B (XauB), less virulent specie, both after growth in vivo. Initially, we performed growth curves of both bacteria on leaves of a common citrus host (Citrus aurantifolia) in order to investigate the dynamics of population growth in vivo and the efficiency of cell recovery by two different methods. For proteomic analysis, intact bacterial cells had their surface proteins labeled with fluorescence (DIGE CyDye Fluor minimal dyes), were then lysed and the total protein extract analyzed by differential gel electrophoresis (2D-DIGE), as standardized in this study. Protein profiles were analyzed by DeCyder 7.0 software and spots differentially expressed (ANOVA p <0.05) were isolated from gels, identified by mass spectrometry and search in protein databases of the annotated genome sequence of the bacteria. Seventy-nine spots from XAC were analyzed and thirty different proteins were identified, of which 10 correspond to known membrane or cell surface proteins: Ton-B dependent receptors and OmpA-related proteins exhibited lower expression in infectious condition, differently of Ferric enterobactin receptors, 60 kDa chaperonin (GroEL) and DnaK which showed higher expression after host interaction. XAC and XauB total extraction analysis by shotgun identified just two XAC proteins. Cell surface proteins with increased in vivo expression in virulent strain (XAC) could provide future targets of biotechnological interest for fighting citrus canker for being possibly related to phytopathogenicity and/or host spectrum.
Siga as redes sociais do SIBi
Contato

Repositório Institucional da UFSCar (RI UFSCar)

Av. Biblioteca Comunitária, Campus São Carlos

Rod. Washington Luís, km 235 - SP-310

São Carlos, SP, 13565-905 (16) 3351-8477

Email: repositorio@ufscar.br

Acessos ao repositorio.ufscar.br

DSpace software copyright © 2002-2026 LYRASIS

  • Acessibilidade
  • Política de Privacidade
  • Termos de Uso
  • Enviar uma Sugestão